Operation to invoke a Harness¶

Description¶

Operation to invoke a Harness.

Usage¶

bedrockagentcore_invoke_harness(harnessArn, qualifier, runtimeSessionId,
  runtimeUserId, traceParent, traceState, traceId, baggage, messages,
  model, systemPrompt, tools, skills, allowedTools, maxIterations,
  maxTokens, timeoutSeconds, actorId)

Arguments¶

harnessArn

[required] The ARN of the harness to invoke.
qualifier

The endpoint name to invoke. If omitted, the DEFAULT endpoint is used.
runtimeSessionId

[required] The session ID for the invocation. Use the same session ID across requests to continue a conversation.
runtimeUserId

An identifier for the end user making the request. This value is passed through to the runtime container.
traceParent

W3C trace context parent header containing version, trace ID, parent span ID, and trace flags.
traceState

W3C trace context state header for vendor-specific trace information.
traceId

Trace ID for maintaining observability through the operation.
baggage

W3C Baggage header for user-defined context propagation. Format: key1=value1,key2=value2
messages

[required] The messages to send to the agent.
model

The model configuration to use for this invocation. If specified, overrides the harness default.
systemPrompt

The system prompt to use for this invocation. If specified, overrides the harness default.
tools

The tools available to the agent for this invocation. If specified, overrides the harness default.
skills

The skills available to the agent for this invocation. If specified, overrides the harness default.
allowedTools

The tools that the agent is allowed to use for this invocation. If specified, overrides the harness default.
maxIterations

The maximum number of iterations the agent loop can execute. If specified, overrides the harness default.
maxTokens

The maximum number of tokens the agent can generate per iteration. If specified, overrides the harness default.
timeoutSeconds

The maximum duration in seconds for the agent loop execution. If specified, overrides the harness default.
actorId

The actor ID for memory operations. Overrides the actor ID configured on the harness.

Value¶

A list with the following syntax:

list(
  stream = list(
    messageStart = list(
      role = "user"|"assistant"
    ),
    contentBlockStart = list(
      contentBlockIndex = 123,
      start = list(
        toolUse = list(
          toolUseId = "string",
          name = "string",
          type = "tool_use"|"server_tool_use"|"mcp_tool_use",
          serverName = "string"
        ),
        toolResult = list(
          toolUseId = "string",
          status = "success"|"error"
        )
      )
    ),
    contentBlockDelta = list(
      contentBlockIndex = 123,
      delta = list(
        text = "string",
        toolUse = list(
          input = "string"
        ),
        toolResult = list(
          list(
            text = "string",
            json = list()
          )
        ),
        reasoningContent = list(
          text = "string",
          redactedContent = raw,
          signature = "string"
        ),
        toolResultMetadata = list(
          metadata = "string"
        )
      )
    ),
    contentBlockStop = list(
      contentBlockIndex = 123
    ),
    messageStop = list(
      stopReason = "end_turn"|"tool_use"|"tool_result"|"max_tokens"|"stop_sequence"|"content_filtered"|"malformed_model_output"|"malformed_tool_use"|"interrupted"|"partial_turn"|"model_context_window_exceeded"|"max_iterations_exceeded"|"max_output_tokens_exceeded"|"timeout_exceeded"|"hook_stopped"
    ),
    metadata = list(
      usage = list(
        inputTokens = 123,
        outputTokens = 123,
        totalTokens = 123,
        cacheReadInputTokens = 123,
        cacheWriteInputTokens = 123
      ),
      metrics = list(
        latencyMs = 123
      )
    ),
    internalServerException = list(
      message = "string"
    ),
    validationException = list(
      message = "string",
      reason = "CannotParse"|"FieldValidationFailed"|"IdempotentParameterMismatchException"|"EventInOtherSession"|"ResourceConflict",
      fieldList = list(
        list(
          name = "string",
          message = "string"
        )
      )
    ),
    runtimeClientError = list(
      message = "string"
    ),
    hookEvent = list(
      hookEventId = "string",
      name = "string",
      type = "before_tool_call"|"after_tool_call"|"before_invocation"|"after_invocation",
      decision = "allow"|"deny",
      reason = "string"
    )
  )
)

Request syntax¶

svc$invoke_harness(
  harnessArn = "string",
  qualifier = "string",
  runtimeSessionId = "string",
  runtimeUserId = "string",
  traceParent = "string",
  traceState = "string",
  traceId = "string",
  baggage = "string",
  messages = list(
    list(
      role = "user"|"assistant",
      content = list(
        list(
          text = "string",
          toolUse = list(
            name = "string",
            toolUseId = "string",
            input = list(),
            type = "tool_use"|"server_tool_use"|"mcp_tool_use",
            serverName = "string"
          ),
          toolResult = list(
            toolUseId = "string",
            content = list(
              list(
                text = "string",
                json = list()
              )
            ),
            status = "success"|"error",
            type = "tool_use"|"server_tool_use"|"mcp_tool_use"
          ),
          reasoningContent = list(
            reasoningText = list(
              text = "string",
              signature = "string"
            ),
            redactedContent = raw
          )
        )
      )
    )
  ),
  model = list(
    bedrockModelConfig = list(
      modelId = "string",
      maxTokens = 123,
      temperature = 123.0,
      topP = 123.0,
      apiFormat = "converse_stream"|"responses"|"chat_completions",
      additionalParams = list()
    ),
    openAiModelConfig = list(
      modelId = "string",
      apiKeyArn = "string",
      apiBase = "string",
      maxTokens = 123,
      temperature = 123.0,
      topP = 123.0,
      apiFormat = "chat_completions"|"responses",
      additionalParams = list()
    ),
    geminiModelConfig = list(
      modelId = "string",
      apiKeyArn = "string",
      maxTokens = 123,
      temperature = 123.0,
      topP = 123.0,
      topK = 123,
      additionalParams = list()
    ),
    liteLlmModelConfig = list(
      modelId = "string",
      apiKeyArn = "string",
      apiBase = "string",
      maxTokens = 123,
      temperature = 123.0,
      topP = 123.0,
      additionalParams = list()
    )
  ),
  systemPrompt = list(
    list(
      text = "string"
    )
  ),
  tools = list(
    list(
      type = "remote_mcp"|"agentcore_browser"|"agentcore_gateway"|"inline_function"|"agentcore_code_interpreter",
      name = "string",
      config = list(
        remoteMcp = list(
          url = "string",
          headers = list(
            "string"
          )
        ),
        agentCoreBrowser = list(
          browserArn = "string"
        ),
        agentCoreGateway = list(
          gatewayArn = "string",
          outboundAuth = list(
            awsIam = list(),
            none = list(),
            oauth = list(
              providerArn = "string",
              scopes = list(
                "string"
              ),
              customParameters = list(
                "string"
              ),
              grantType = "CLIENT_CREDENTIALS"|"AUTHORIZATION_CODE"|"TOKEN_EXCHANGE",
              defaultReturnUrl = "string"
            )
          )
        ),
        inlineFunction = list(
          description = "string",
          inputSchema = list()
        ),
        agentCoreCodeInterpreter = list(
          codeInterpreterArn = "string"
        )
      )
    )
  ),
  skills = list(
    list(
      path = "string",
      s3 = list(
        uri = "string"
      ),
      git = list(
        url = "string",
        path = "string",
        auth = list(
          credentialArn = "string",
          username = "string"
        )
      ),
      awsSkills = list(
        paths = list(
          "string"
        )
      )
    )
  ),
  allowedTools = list(
    "string"
  ),
  maxIterations = 123,
  maxTokens = 123,
  timeoutSeconds = 123,
  actorId = "string"
)